CONTROL SYSTEM FOR STEERING AN AGRICULTURAL MACHINE

DE502022007575D1Active Publication Date: 2026-04-23CLAAS E SYSTEMS GMBH
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
DE · DE
Patent Type
Patents
Current Assignee / Owner
CLAAS E SYSTEMS GMBH
Filing Date
2022-06-24
Publication Date
2026-04-23

AI Technical Summary

Technical Problem

Operators of agricultural machinery face challenges in navigating and maneuvering large machines with attached or towed devices in difficult situations, such as parking, coupling, and maneuvering around obstacles, which require significant attention and effort, increasing the risk of accidents.

Method used

A control system for agricultural machines that includes a control unit, storage unit, and processing unit to record and execute steering and driving sequences learned from operator maneuvers, allowing for automated execution of these maneuvers in similar situations, supported by sensor data and learning modes.

Benefits of technology

Reduces operator burden in repetitive maneuvers by providing automated support, minimizing accidents and enhancing safety and efficiency in navigating and maneuvering large agricultural machines.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader
Need to check novelty before this filing date? Find Prior Art

Description

[0001] The present invention relates to a control system for steering an agricultural machine according to the preamble of claim 1. The invention further relates to a method for operating a control system for steering an agricultural machine according to the preamble of claim 12.

[0002] From WO2020 / 061362 A1, a system for the autonomous reversing of a tractor and a method for maneuvering a tractor in reverse to couple to a trailer using a coupling assistance system are known. An operator selects a trailer to be coupled, which has been identified by a camera mounted on the tractor. A path planning system determines the route to be traveled, with a starting point at the tractor and an endpoint at the trailer. The tractor autonomously travels the calculated route until it reaches the trailer. During reversing, the route is continuously monitored for obstacles along the calculated path, and the route is adjusted as necessary.

[0003] WO 2020 / 076427 A1 relates to a steering system for steering an agricultural machine and a towed implement attached to it, according to the preamble of claim 1. For setting down or parking the implement, a target point for placement within a designated area and a path to this target point, to be traveled by reversing the machine, are defined. The steering system calculates steering commands to automatically steer the implement along this path.

[0004] Another control system and method for steering an agricultural machine according to the preamble of claim 1 is known from German patent application DE 10 2018 210379 A1. This document describes a control system for steering an agricultural machine with at least one device arranged thereon and / or pulled by it, wherein the control system comprises a control unit with a storage unit for retrievable storage of data and a processing unit configured to process the data stored in the storage unit and to generate steering signals from it for controlling a steering device of the machine.The control system is designed to execute a learning mode in order to record data on steering and driving behavior, which is based on at least one steering maneuver manually controlled by an operator, and to store this data in the memory unit as at least one executable control sequence.

[0005] The invention is based on the objective of further developing a control system and a method of the type mentioned above, whereby an operator of the agricultural machinery is relieved of the burden of navigating and maneuvering with mounted and / or towed devices in recurring situations.

[0006] The problem is solved by a control system for steering an agricultural machine with the features of independent claim 1 and a method for operating a control system for steering an agricultural machine with the features of dependent claim 12.

[0007] Advantageous embodiments and further developments can be found in the dependent claims.

[0008] According to claim 1, a control system for steering an agricultural machine is proposed, comprising at least one device arranged thereon and / or towed by it. The control system includes a control unit, a storage unit for retrievable data storage, and a processing unit configured to process the data stored in the storage unit and generate steering signals from it for controlling a steering device of the machine. The invention provides that the control system is configured to execute a manually initiated learning mode in order to acquire data on steering and driving behavior resulting from at least one steering maneuver manually controlled by an operator and to store this data in the storage unit as at least one executable control sequence.

[0009] The invention is based on the idea that the operator receives support during difficult driving maneuvers, particularly those that are repetitive in daily work, with the agricultural machine and the attached and / or towed device, in order to relieve the operator. Steering maneuvers include maneuvering and / or shunting in difficult situations such as parking, approaching a connection to a container for receiving liquid, coupling and uncoupling the towed device, picking up and setting down the device, passing through passages whose width only slightly exceeds the width of the agricultural machine and the attached and / or towed device, and / or driving around obstacles.By manually initiating the learning mode, the operator records all steering maneuvers performed manually during maneuvering and / or parking and stores them as an executable control sequence in the memory unit. The term "steering maneuver" is to be interpreted broadly and includes not only the operation of a steering device, such as a steering wheel, but also any related actions in driving operations, such as acceleration, deceleration, and / or gear changes. During learning mode, a start and end point of the journey can be defined, even automatically, and stored as data in the recorded control sequence.

[0010] The term "attached device" refers to implements that are directly attached to a mounting device of the working machine and are transported with the working machine at a distance from the ground, provided they do not have ground-penetrating devices such as support wheels for transport purposes, for example, a corn header or a plow. The term "trailed device" refers in particular to trailers and implements with ground-penetrating devices, for example, a baler.

[0011] Furthermore, the control system may be configured to detect and record all operator-initiated and machine-executed sequences that occur during the learning mode and that go beyond the execution of steering maneuvers. The control system may be configured to chronologically assign these sequences to the control history. Operator-initiated and machine-executed sequences include, in particular, the activation of the machine's actuators during the learning mode, which are used to operate devices mounted on the machine. Machine actuators may, for example, be individual hydraulic cylinders used to operate a front or rear linkage or a feed channel, for instance, to attach or detach an implement or to pick up or remove a front attachment.

[0012] In particular, the control system can be configured to retrieve the executable control sequences stored in the memory unit at a later time, depending on the event. Event-dependent retrieval can involve manually retrieving a control sequence selected by the operator. It can also result from the automated detection of a current operating situation.

[0013] Preferably, the control system can be configured to trigger an executable control sequence manually by a request from the operator and / or automatically depending on an operating situation of the agricultural machinery detected by the control system.

[0014] According to the invention, the control system is configured to determine and execute the corresponding control sequence stored in the memory unit based on the specification of a start point and an end point. This sequence comprises steering maneuvers performed between this start point and this end point during a previously executed learning mode. For example, the corresponding control sequence can be selected by the control system through manual input of the start point and end point using an input / output unit of the control system.

[0015] According to a preferred further development approach, the control system can be configured to compare and adapt steering profiles recorded between the same start and end points during repeatedly executed learning modes. By comparing the recorded steering profiles, individual partial steering maneuvers can be identified and combined into an adapted steering profile to optimize the steering maneuver to be executed by the control system.

[0016] In particular, the control system can include at least one sensor array, and in particular several sensor arrays, configured to provide data from which the steering and driving behavior during the learning mode can be determined. For example, a speed sensor associated with the powertrain can provide data relating to the acceleration and deceleration of the agricultural machine during the learning mode. Other powertrain data that can be sensorily recorded can include torque and the selected gear. Similarly, a steering angle sensor can be assigned to the steering device to record steering movements performed by the operator. Furthermore, a position tracking sensor can be provided on the machine so that the starting point, end point, and the distance traveled between them can be recorded.

[0017] For this purpose, the control system can be configured to correlate the data from at least one sensor array with the steering maneuver manually performed by the operator, evaluate it, and store it in the control sequence to be saved. In this way, all movements of the machine and / or device during navigation and maneuvering can be temporally assigned to actions of the operator that are necessary for navigation and maneuvering.

[0018] According to further training, the control system can be configured to save the steering and driving behavior performed during learning mode, depending on the type and / or category of the device attached to and / or towed by the machine. The type can be categorized as either towed or attached to the machine, as well as by their intended use. Categorization by different types can be based, for example, on manufacturers and model designations.

[0019] In particular, at least one optical sensor device can be arranged on the working machine as a sensor arrangement, which is configured for environmental detection. Preferably, at least two optical sensor devices are provided, which can detect the environment of the working machine at least in the forward and reverse directions. Furthermore, optical sensor devices can be provided that detect the lateral environment of the working machine and the device attached to or towed by it.

[0020] In particular, the optical sensor device, which is designed to detect the surroundings in the reverse direction, can detect the coupling point between the machine and the towed device. The towed device generally has a drawbar or similar component with which it is coupled to the machine. This allows the corresponding reactions of the towed device to be correlated with the steering movements of the machine. For example, operator errors during steering and subsequent corrective steering movements can be detected and analyzed.

[0021] According to another aspect, the control system can be configured to recognize an upcoming operating situation based on an evaluation of the image data provided by at least one optical sensor device. The steering and driving behavior for this situation has already been recorded and stored during a learning mode. An operating situation could, for example, be moving the machine into a parking position on a farmyard. Another operating situation could be picking up or coupling the attachment when the machine approaches the attachment and the attachment is recognized as such, for which at least one corresponding control profile is stored in the memory unit.Recognizing an operating situation allows the control system to output at least a suggested control sequence to the input / output unit, based on how a comparable or identical operating situation was handled, i.e., which steering maneuvers were recorded in a previously executed steering mode. This recognition can also occur independently of the at least one optical sensor device or with the support of the position tracking sensor on the machine. For example, an existing parking position of the machine can be detected, which the machine is to leave when it is started up.

[0022] Furthermore, the control system can be configured to make at least one suggestion, depending on the detected upcoming operating situation, as to which steering maneuver can be executed in the detected operating situation based on the control profiles for steering and driving behavior stored in the memory unit.

[0023] In particular, the control system can be configured to execute a control sequence retrieved from the storage unit in reverse order.

[0024] The problem initially set out is also solved by a method for operating a control system for steering an agricultural machine with the features of dependent claim 12.

[0025] According to claim 12, a method for operating a control system for steering an agricultural machine with at least one device arranged thereon and / or towed by it is proposed, wherein the control system comprises a control unit, a storage unit for retrievable storage of data, and a processing unit by which data stored in the storage unit are processed in order to generate steering signals with which a steering device of the machine is controlled, wherein a manually initiated learning mode is executed in which data of a steering and driving behavior resulting from at least one steering maneuver manually controlled by an operator are recorded and stored in the storage unit as an executable control sequence, and based on the determination of a start point and an end point, the corresponding control sequence stored in the storage unit is generated.which includes steering maneuvers performed between this starting point and this endpoint in a previously executed learning mode, is determined and executed.

[0026] Reference may be made to the advantages of the control system according to the invention.

[0027] Depending on a detected operating situation of the agricultural machine or a manual trigger by the operator, an executable control sequence stored in the memory unit can be retrieved and executed by the control system to steer the agricultural machine.

[0028] Preferably, all processes initiated by the operator and executed by the machine during a learning mode, which go beyond steering, can be recorded and assigned to the control sequence as driving behavior. The term "driving behavior" includes not only steering maneuvers but also additional processes performed by the machine that are directly related to the steering maneuvers being recorded, either temporally or spatially. For example, a process to be recorded during a steering maneuver could be the coupling or uncoupling of an attachment or the picking up or dropping of a front attachment. Coupling or uncoupling an attachment or picking up or dropping a front attachment involves the targeted control of actuators on the machine, which is initiated by the operator.

[0029] According to further training, at least one control sequence can be suggested for execution by the control system based on the situation. This can be done, for example, based on environmental detection and / or by inputting a start and end point, for which previously recorded and saved control sequences have been stored.

[0030] The present invention is explained in more detail below with reference to an embodiment illustrated in the drawings.

[0031] They show: Fig. 1 a schematic representation of a control system for steering an agricultural machine; Fig. 2a) schematically a flowchart of an exemplary sequence of a learning mode, b) schematically a flowchart of an exemplary sequence of an extended learning mode, and c) schematically a flowchart of an exemplary sequence of an execution of a control sequence recorded in learning mode; Fig. 3a) to d) a schematic representation of a parking procedure of the agricultural machine with a transport trailer attached; and Fig. 4a) to d) a schematic representation of the parking procedure according to the Fig. 3 a) bis 3 d) taking into account an obstacle.

[0032] In Fig. 1 The basic structure of a proposed control system 1 for steering an agricultural machine 2 and a device 3 attached to and / or towed by it is schematically illustrated using an exemplary embodiment. The agricultural machine 2 can be a self-propelled harvester or a tractor. The device 3 can be an attachment or implement that is picked up by a receiving device on the machine 2, for example, a feed chute or a three-point linkage, and attached to the machine 2. The device 3 can also be a towed harvester, for example, a baler, or a trailer that is coupled to a coupling device on the machine 2. The coupling device on the machine 2 can, for example, be a three-point linkage or a drawbar.

[0033] The control system 1 comprises a control unit 4 with a storage unit 5 for retrievable data storage and a processing unit 6, which is configured to process the data stored in the storage unit 5 and to generate steering signals LS from it for controlling a steering device 8 of the working machine 2. Furthermore, the control unit 4 has an input / output unit 7.

[0034] The control system 1 comprises at least one sensor arrangement 10a, in particular several sensor arrangements 10b, 10c, 10d, 10e, which are configured to provide data from the working machine 1 and / or the device 3 arranged on it and / or pulled by it. Depending on the design and operation of the sensor arrangements 10b, 10c, 10d, 10e, the data provided by them includes, among other things, measurement signals, image data, and the like.

[0035] A sensor arrangement 10a of the working machine 2 can be assigned to a drive train 9 of the working machine 2, which consists, among other things, of a drive motor and a gearbox. The sensor arrangement 10a assigned to the drive train 9 can detect rotational speed and / or torque as well as the selected gear, so that the acceleration and deceleration of the working machine 2 and its direction of travel can be determined. In addition, a sensor arrangement (not shown) can be provided that detects the pedal position of the acceleration and deceleration pedals.

[0036] A further sensor arrangement 10b can be assigned to the steering device 8 to detect a steering angle. Alternatively, the sensor arrangement 10b can be assigned to a steering axle of the working machine 2, which is controlled by the steering device 8 to execute steering movements.

[0037] As a further sensor arrangement 10c, at least one optical sensor device can be arranged on the working machine 1, which is configured for environmental detection. Preferably, at least two sensor arrangements 10c designed as optical sensor devices are provided, which can detect the environment of the working machine 2 at least in the forward and reverse directions. For this purpose, one sensor arrangement 10c can be arranged at the front and another sensor arrangement 10c at the rear of the working machine 2. Furthermore, sensor arrangements 10c designed as optical sensor devices can be provided, which detect the lateral environment of the working machine 2 and the device 3 attached to and / or towed by it.

[0038] At least on the working machine 2, a sensor arrangement 10d comprising a position tracking sensor is arranged, which enables the recording and saving of a traveled distance based on a large number of waypoints.

[0039] The optical sensor arrangement 10c, located in the rear area of ​​the machine 2, can also be used to detect the steering angle of a drawbar 3a of the device 3. Alternatively or additionally, a sensor arrangement 10e, configured as an articulation angle sensor or steering angle sensor, can be assigned to the drawbar 3a. The sensor arrangement 10c, located in the rear area of ​​the machine 2, allows the device 3 to detect reactions to steering movements transmitted by the machine 2.

[0040] The signals provided by the sensor arrangements 10a, ..., 10e are supplied to the control unit 4 for processing by algorithms stored in the storage unit 5 and executed by the computing unit 6.

[0041] Today's agricultural machinery 2 is becoming increasingly larger and, due to its dimensions, increasingly difficult for the operator to oversee. A large implement 3, such as a harrow, sprayer, plow, etc., with a significant reach, further exacerbates this situation. While operators are generally able to control the machine 2 with the attached implement 3 safely, this requires considerable effort in every critical situation, such as reversing, due to the need for heightened attention. Critical situations include parking or stopping the machine 2 and / or the implement 3, accessing a connection for operating fluids or supplies, attaching or detaching the implement 3, or maneuvering around obstacles, especially under challenging conditions such as confined spaces.

[0042] To minimize the risk of accidents during navigation and maneuvering, the operator of the work machine 2 should be supported by the control system 1 in vehicle guidance in recurring similar difficult situations, especially during navigation and maneuvering.

[0043] The representation in Fig. 2a Figure 1 schematically shows a flowchart of an exemplary process in a learning mode. Fig. 2 b) The diagram schematically shows a flowchart of an exemplary process in an extended learning mode, and in Fig. 2c ) The representation schematically shows a flowchart of an exemplary sequence of the execution of a control sequence recorded in a learning mode and stored in memory unit 5 11.

[0044] With reference to Fig. 2a The execution of the learning mode is explained first. Using the input / output unit 7, the operator of the machine 2 can manually initiate the learning mode. In the first procedure step 12, the operator calls up the learning mode.

[0045] In the subsequent process step 13, all steering maneuvers performed by the operator during the active learning mode are automatically recorded. For this purpose, the signals provided by the existing sensor arrangements 10a, ..., 10e are evaluated by the processing unit 6 to generate a reproducible control sequence 11, which includes all actions performed by the operator during the learning mode, insofar as these relate to steering and actuating, i.e., controlling actuators, the machine 2 or the device 3. The control system 1 is configured to correlate and evaluate the data from the at least one sensor arrangement 10a, ..., 10e with the steering maneuver manually performed by the operator.

[0046] In process step 14, the recording of the executed steering maneuvers and the associated data is terminated. Recording can be terminated by the operator using the input / output unit 7. The operator thereby confirms the path traveled by the machine 2. Alternatively, the termination of recording can be detected based on data provided by one of the sensor arrangements 10a, ..., 10e. For example, the sensor arrangement 10d, which includes the position tracking sensor, can determine when a predefined endpoint or target point has been reached, thus triggering the termination of recording. Another option is to switch off the drive motor of the machine 2, which automatically ends the recording.

[0047] In the subsequent process step 15, the steering maneuvers and actions performed during navigation and shunting are saved in the storage unit 5 as a control sequence 11.

[0048] In Fig. 2b Figure 1 shows a schematic and exemplary flowchart of an extended learning mode. After manually calling up the learning mode in the first process step 12, the system can optionally proceed to process step 12a, in which the operator selects a previously saved control sequence 11. Control sequence 11 was recorded earlier for the same operating situation. Subsequently, process steps 13 and 14, which were explained above, are executed.

[0049] Procedure step 14 is followed by a further optional procedure step 14a, which is executed depending on the execution of procedure step 12. In procedure step 14a, the corresponding steering maneuvers stored in the previous control profile 11 are adjusted based on the recording of differing steering maneuvers. If a difference is detected by comparing the steering maneuvers stored in control profile 11 with the newly recorded steering maneuvers, and this difference represents an improvement in the sense of optimization—for example, a reduction in steering movements or a shortening of the distance to be traveled—an adjusted control profile 11a is saved instead of control profile 11.The occurrence of a change, for example due to a change in the route because of a newly added obstacle or the like, which necessitates at least a partial adjustment of steering maneuvers, also leads to an adapted steering profile 11a being saved instead of the steering profile 11.

[0050] During the process according to Fig. 2a ) a direct imitation of the driving maneuvers performed by the operator during learning mode, the extended procedure according to Fig. 2 b) Repeated execution and recording of the same operating situation in several successive learning modes allows the operator to train the machine. This enables the machine to learn a model of how to resolve a situation and adapt it based on each repeated training session.

[0051] Furthermore, it is conceivable that the control system 1 could be configured to determine an initial state of the machine 2, including a starting position, and a final state of the machine 2, including a final position, relative to the initial state and starting position. The path to get from the initial state and starting position to the final state and final position can be automatically determined by the machine 2 using planning algorithms. For example, an optimal path to the target position can be planned when parking.For example, the machine 2 can be moved from a park position of the machine 2 as the starting point, in which the machine 2 is in its initial state without device 3, to a rejected endpoint, which corresponds to the location of a device 3 to be picked up or coupled, by means of a control sequence 11 determined by the planning algorithm, which corresponds to the location of a device 3 to be picked up or coupled, wherein the picked up or coupled device 3 represents the final state.

[0052] With reference to Fig. 2c ) The retrieval of a stored tax history 11 and the triggering of the execution of the tax history 11 are explained.

[0053] In step 16, a stored control sequence 11 is retrieved depending on an event. An event can be a manual retrieval of the stored control sequence 11. Alternatively, a stored control sequence 11 can be retrieved when a specific start and end point are entered by the operator using the input / output unit 7, which are assigned to this control sequence 11 during a previously performed learning mode. Event-dependent retrieval can also be due to the automated detection of a current operating situation.

[0054] The control system 1 can be configured to recognize an upcoming operating situation based on an evaluation of the image data provided by at least one optical sensor device, for which the steering and driving behavior has already been recorded and stored during a learning mode. An operating situation could, for example, be moving the machine 2 into a parking position in a yard. Another operating situation could be picking up or coupling the device 3 when the machine 2 approaches a device and the device 3 is recognized as such, for which at least one corresponding control sequence 11 is stored in the memory unit 5.

[0055] Furthermore, the detection of an operating situation enables the control system 1 to output at least one suggestion for a control sequence 11 to the input / output unit 7, indicating which steering maneuvers have already been recorded using an executed steering mode in a comparable or identical operating situation. The detection of an operating situation can also occur independently of the at least one optical sensor device as sensor arrangement 10c and / or supported by the position tracking sensor of sensor arrangement 10d on the machine 2. For example, an existing parking position of the machine 2 can be detected, which the machine 2 is to leave when it is started up.

[0056] According to further training, the control system 1 can be configured to save the steering and driving behavior performed during learning mode, depending on the type and / or category of the device 3 attached to and / or towed by the working machine 2. The type can be categorized as towed devices 3, such as a transport wagon or baler, and devices 3 attached to and carried by the working machine 2, such as a tillage implement, as well as their intended use. Categorization by different types can be based, for example, on manufacturers and model designations. Additionally, the detected environment can be used as a further criterion. This includes, in particular, different locations on the farmyard.

[0057] The control system 1 is configured to execute a control sequence 11 retrieved from the storage unit 5 in reverse order. For example, the control sequence 11 of a recorded parking maneuver can be executed in reverse order, so that the work machine 2 independently performs the parking maneuver according to the selected control sequence 11.

[0058] In Fig. 3 Figure 2 is a schematic representation of the sequence of a parking process of the agricultural work machine 2 with a transport wagon 21 attached to it as device 3.

[0059] Fig. 3 a) Figure 2 shows the work machine 2 and the transport vehicle 21 at their starting point. A steering wheel position 19 of the steering device 8, which is designed as a steering wheel, is shown as the starting position, corresponding to driving straight ahead. A parking area 18 forms the endpoint to be reached at the end of the parking process. The parking area 18 can be a free, flat surface or be limited by lateral and / or rear boundaries, e.g., walls or other devices 3, which makes maneuvering and parking particularly complex for the operator. In order to have the parking process automatically executed again at a later time by the control system 1, the operator starts the learning mode according to procedure step 12 of the Fig. 2 a) oder 2 b ). The operator then begins to perform the necessary steering maneuvers, which, as described above, are recorded in procedure step 13.

[0060] In Fig. 3 b) In addition to the changed steering wheel position 19, which indicates a turn, a direction of travel, here a reverse direction, and a direction of acceleration are also illustrated by an arrow 20. The in Fig. 3b The steering wheel position shown (19) differs from the one shown in Fig. 3a ) shown and is recorded according to procedure step 13 during the execution of the learning mode.

[0061] In Fig. 3 c) The transport vehicle 21 is shown immediately before reaching parking area 18.

[0062] In Fig. 3 d) The work machine 2 and the transport vehicle 21 have reached their final state. The transport vehicle 21 is located at its endpoint in parking area 18. Symbol 22 indicates that it has come to a standstill. The indicated steering wheel position 19 corresponds to driving straight ahead. Upon reaching the endpoint, the recording is terminated according to procedure step 14. In step 15, the steering maneuvers performed in connection with reaching the endpoint are saved as a control sequence 11. For exiting the parking space and subsequent re-parking, the operator can, for example, manually recall the control sequence 11 and have the control system 1 execute it automatically for exiting the parking space, then in reverse order, or for re-parking.

[0063] In the Fig. 4 a) bis 4 d ) is the one based on the Fig. 3 a) bis d) The previously described procedure for reversing the transport vehicle 21 into parking space is described. The required steering maneuvers differ due to an obstacle 22 in the maneuvering area of ​​parking space 18. Thus, the work machine 2 and the transport vehicle 21 must first reverse so far that the work machine 2, as described in the previous section, is positioned correctly. Fig. 4 b) This illustrates how the vehicle can swivel in front of obstacle 22 to align itself and the transport vehicle 21 perpendicularly to the parking area 18. The steering maneuvers according to the Fig. 4 c) und Fig. 4 d) correspond to those according to the Fig. 3 c) und Fig. 3 d) . During the learning mode, in the operating situation shown with the obstacle 22, its position and the additional steering maneuvers required by the operator during the learning mode are recorded and saved in a control history 11.

[0064] Furthermore, it may be possible to record and capture all processes initiated by the operator and executed by the machine 2 during the learning mode that go beyond the execution of steering maneuvers and are temporally and / or spatially related. The control system 1 may be configured to chronologically assign these processes to the control sequence 11, 11a. Processes initiated by the operator and executed by the machine 1 include, in particular, the actuation of actuators of the machine 2 during the learning mode, which serve to operate devices arranged on the machine 2. Actuators of the machine 2 may, for example, be individual hydraulic cylinders with which a front or rear linkage or a feed channel is operated, or a hydraulic system for, for example, attaching or detaching an implement or picking up or removing a front attachment. Reference symbol list

[0065] 1 Control system 2 Working machine 3 Device 3a Drawbar 4 Control unit 5 Storage unit 6 Computing unit 7 Input / output unit 8 Steering device 9 Drive train 10a Sensor arrangement 10b Sensor arrangement 10c Sensor arrangement 10d Sensor arrangement 10e Sensor arrangement 11 Control sequence 11a Control sequence 12 Process step 12a Process step 13 Process step 14 Process step 14a Process step 15 Process step 16 Process step 17 Process step 18 Parking area 19 Steering wheel position 20 Arrow "Direction of travel and acceleration" 21 Transport trolley 22 Symbol "Standstill"

Claims

1. Control system (1) for steering an agricultural work machine (2) having at least one apparatus (3) arranged thereon and / or pulled thereby, wherein the control system (1) comprises a control unit (4) with a storage unit (5) for retrievably storing data and a computing unit (6) which is configured to process the data stored in the storage unit (5) and to generate steering signals (LS) therefrom for controlling a steering device (8) of the work machine (2), wherein the control system (1) is configured to execute a learning mode in order to record in the learning mode data relating to a steering and driving behaviour which goes back to at least one steering manoeuvre manually controlled by an operator, and to retrievably store said data as at least one executable control profile (11, 11a) in the storage unit (5), characterized in that the learning mode is a manually initiated learning mode, and in that the control system (1) is configured to determine the corresponding control profile (11, 11a) stored in the storage unit (5) based on a determination of a starting point and an end point, which control profile comprises steering manoeuvres performed in a previously executed learning mode between that starting point and that end point, and to execute said control profile.

2. Control system (1) according to Claim 1, characterized in that the control system (1) is designed to retrieve the executable control profiles (11, 11a) stored in the storage unit (5) at a later time depending on the event.

3. Control system (1) according to Claim 1 or 2, characterized in that the control system (1) is configured to trigger an executable control profile (11, 11a) manually by a retrieval by the operator and / or automatically on the basis of an operating situation of the agricultural work machine (2) that is detected by the control system (1).

4. Control system (1) according to one of Claims 1 to 3, characterized in that the control system (1) is configured to compare and adapt control profiles (11, 11a) which were recorded during repeatedly executed learning modes between the same starting point and the same end point.

5. Control system (1) according to one of Claims 1 to 4, characterized in that the control system (1) comprises at least one sensor arrangement (10a, 10b, 10c, 10d, 10e), in particular a plurality of sensor arrangements (10a, 10b, 10c, 10d, 10e), configured to provide data that can be used to determine the steering and driving behaviour during the learning mode.

6. Control system (1) according to Claim 5, characterized in that the control system (1) is configured to put the data from the at least one sensor arrangement (10a, 10b, 10c, 10d, 10e) in a temporal relationship with the steering manoeuvre manually performed by the operator, to evaluate them and to store them in the control profile (11, 11a) to be stored.

7. Control system (1) according to one of the preceding claims, characterized in that the control system (1) is configured to store the steering and driving behaviour carried out during the learning mode on the basis of the kind and / or type of apparatus (3) arranged on the work machine (2) and / or pulled.

8. Control system (1) according to either of Claims 6 and 7, characterized in that, as a sensor arrangement (10a, 10b, 10c, 10d, 10e), at least one optical sensor device (10c) is arranged on the work machine (2) and is configured to detect the environment.

9. Control system (1) according to Claim 8, characterized in that the control system (1) is configured to recognize an impending operating situation, for which the steering and driving behaviour has already been recorded and stored during a learning mode, on the basis of an evaluation of the image data provided by the at least one optical sensor device (10c).

10. Control system (1) according to Claim 9, characterized in that the control system (1) is configured to present at least one suggestion, on the basis of the recognized impending operating situation, as regards which steering manoeuvre can be performed on the basis of the control profiles (11, 11a) for steering and driving behaviours stored in the storage unit (5).

11. Control system (1) according to one of the preceding claims, characterized in that the control system (1) is configured to execute a control profile (11, 11a) retrieved from the storage unit (5) in reverse order.

12. Method for operating a control system (1) for steering an agricultural work machine (2) having at least one apparatus (3) arranged thereon and / or pulled thereby, wherein the control system (1) comprises a control unit (4), a storage unit (5) for retrievably storing data and a computing unit (6) that is used to process data stored in the storage unit (5) in order to generate steering signals (LS) therefrom, with which a steering device (8) of the work machine (2) is controlled, characterized in that that a manually initiated learning mode is executed in which data relating to a steering and driving behaviour, which goes back to at least one steering manoeuvre manually controlled by an operator, are recorded and retrievably stored as an executable control profile (11, 11a) in the storage unit (5), and in that the corresponding control profile (11, 11a) stored in the storage unit (5) is determined based on a determination of a starting point and an end point, which control profile comprises steering manoeuvres performed in a previously executed learning mode between that starting point and that end point, and is executed.

13. Method according to Claim 12, characterized in that, on the basis of a recognized operating situation of the agricultural work machine (2) or manual triggering by the operator, an executable control profile (11, 11a) stored in the storage unit (5) is retrieved and is executed by the control system (1) for steering the agricultural work machine (2).

14. Method according to Claim 12 or 13, characterized in that all processes initiated by the operator during a steering manoeuvre to be recorded and carried out by the work machine (2), which go beyond steering, are recorded as driving behaviour and assigned to the control profile (11, 11a).